microsoft/onnxruntime

Pull Requests Commits

[Cuda] support 8 bits in MatMulNBits (#24509)

tianleiwu committed 279 days ago

Verified 3a7c8b3a

webgpu: fix conv error when using MatMulNaiveProgram (#24496)

xhcao committed 279 days ago

Verified 1e118d6a

webgpu: fix InstanceNorm errors (#24514)

xhcao committed 279 days ago

Verified e9bb150d

[Native WebGPU] Fixed Conv2dMM and MatMul issues related indexing, hint, etc. (#24527)

satyajandhyala committed 279 days ago

Verified ef774358

[webgpu] Fix bug in 1D dispatch workgroups (#24519)

qjia7 committed 279 days ago

Verified 5c014e2e

Fix segfault when re-loading an EP library (#24539)

adrianlizarraga committed 279 days ago

Verified 0e407d7b

Fix MatMulScaleFusion handling of scales with leading dimensions. (#24492)

edgchen1 committed 279 days ago

Verified 16996d9f

Remove non-constant input check from ConvActivationFusion (#24525)

edgchen1 committed 279 days ago

Verified cc72c3a9

AutoEP C++ API wrappers (#24501)

skottmckay committed 279 days ago

Verified c7a28aa9

[QDQ Optimizer] Update WeightBiasQuantization to skip Conv/Gemm if downstream node is not QuantizeLinear (#24537)

adrianlizarraga committed 279 days ago

Verified 173a11a4

[webgpu] fix program variable data type and revise ProgramInput (#24533)

fs-eire committed 279 days ago

Verified 19f7ef31

[WebNN] Fix bug in RotaryEmbedding (#24516)

Honry committed 280 days ago

Verified 1de5af84

[x86] matmulnbit x64 kernel for 8bits (#24491)

fajin-corp committed 280 days ago

Verified 7801c513

[QNN-EP] Support for Upsample operator (#24265)

zhaoxul-qti committed 280 days ago

Verified c162821a

NV TensorRT RTX EP - initial commit (#24456)

ankan-ban committed 280 days ago

Verified 2a09f271

bump version from 1.22 to 1.23 (#24506)

vraspar committed 280 days ago

Verified 6c8cb6a6

[WebNN EP] Support MultiHeadAttention(MHA) (#24079)

peishenyan committed 280 days ago

Verified df581e1c

[NativeWebGPU] ConvTranspose and other minor fixes (#24503)

satyajandhyala committed 281 days ago

Verified 8602d04d

Add `python_version < "3.13"` for onnxscript dependency. (#24510)

edgchen1 committed 281 days ago

Verified 8d86dae0

Fix #24130 by adding an empty dependency group to the code that generates Microsoft.ML.OnnxRuntime.nuspec (#24131)

benjamin-hodgson committed 281 days ago

Verified d1ac97e4

[WebNN] Always execute decomposed *SimplifiedLayerNormalization in FP32 (#24437)

Honry committed 281 days ago

Verified 1c6d07f9

Revert "Remove build-nuget from dml-vs-2022.yml" (#24504)

adrianlizarraga committed 281 days ago

Verified 1e3a7dbe

Fix an issue in wasm nortti build and add minimal build support for vcpkg (#24012)

Changming Sun committed 281 days ago

Verified 84cd06c0

Add workflow_dispatch event to GitHub Actions workflows so we can run them manually. (#24505)

edgchen1 committed 282 days ago

Verified 727af017

update test case for test_convtranspose_autopad_same (#24477)

fs-eire committed 282 days ago

Verified 81bb0137

[QNN EP] Update Android QNN package to use QNN SDK 2.33.0 (available on Maven) (#24494)

adrianlizarraga committed 282 days ago

Verified 72f5bad1

[WebGPU EP] Implements Depth-To-Space Operator (#24489)

prathikr committed 282 days ago

Verified fcf2c588

Remove deprecated onnx.mapping usage (#24484)

justinchuby committed 282 days ago

Verified 851d2801

[QNN EP] Remove unnecessary std::move to fix warning/error (#24493)

adrianlizarraga committed 282 days ago

Verified b0e68bde

[QNN EP] Exclude QnnGpu.dll from Windows x64 NuGet (#24487)

adrianlizarraga committed 282 days ago

Verified 5babe2c2

Newer Older